基于混合人工蜂群和人工鱼群优化的LSSVM脉动风速预测

摘要

考虑人工蜂群(ABC)和人工鱼群(AFS)算法的各自优势,提出混合智能算法(ABC+ AFS)优化选择最小二乘支持向量机(LSSVM)参数的方法,以提高其脉动风速预测模型的性能.AFS算法有较强的全局寻优能力,混合智能算法以AFS算法中的人工鱼寻优方式代替ABC算法中的引领蜂寻优方式,克服ABC算法易陷入局部最优的问题.同时,ABC算法中的正负反馈机制可以克服AFS算法的后期盲目寻优、收敛速度下降的问题.运用基于混合ABC、AFS优化的LSSVM对脉动风速进行了预测,并与基于ABC、AFS和粒子群(PSO)算法优化的LSSVM脉动风速预测结果进行了比较.数值结果表明,基于混合ABC+ AFS优化的LSSVM脉动风速预测模型有更好性能,具有工程应用前景.%Considering advantages of the artificial bee colony (ABC)and the artificial fish swarm (AFS)algorithms,a hybrid intelligent optimization algorithm called ABC + AFS algorithm was used to optimize parameters of the least square support vector machine (LSSVM) in order to improve its performance of predicting fluctuating wind velocity.Due to AFS algorithm having a better ability of skipping over local optima,the searching optima mode of artificial bees in ABC algorithm was replaced with that of artificial fishes in AFS algorithm to overcome problems of ABC algorithm being easily to fall into local optima.Concurrently,the positive and negative feedback mechanism in ABC algorithm was used to overcome problems of blindly searching optima and convergent speed dropping in the late period of AFS algorithm.LSSVM based on the ABC + AFS algorithm was used to predict fluctuating wind velocity.The results were compared with those using LSSVM based on ABC,AFS,and PSO algorithms,respectively.The numerical results showed that LSSVM based on the ABC + AFS algorithm has a better performance of predicting fluctuating wind velocity and a bright prospect of engineering application.